Are Amazon Ads Worth It for Authors? One Account, Three Years, Honest Math
InteliAds Team
Amazon KDP PPC
Short answer: Amazon Ads are worth it when your net royalty divided by your list price is higher than the ACOS you can hold, and your book page converts. They are not worth it when the royalty ceiling is under about 20%, when the page does not convert, or when nobody is removing the keywords that never sell. One real KDP account shows both outcomes at the same time.
The receipt
| Whole account, Oct 2023 to Oct 2026 | |
|---|---|
| Ad spend | $48,995.43 |
| Orders attributed to ads | 11,505 |
| Cost per order | $4.26 |
| ACOS | 25.5% |
| October 2026 royalties (paperback, Kindle, KU) | $3,896.07 |
| October 2026 ad spend | $2,329.28 |
| October 2026 net after ads | $1,566.79 |
So yes: over three years, 11,505 orders at $4.26 each on books that net more than that, and a recent month that kept $1,566.79 after the ad bill. For this account the ads are worth it. The rest of this article is about why that is not automatically true for yours.
When they were not worth it, on the same account
Filter the same account's keywords to the ones that took money and never produced an order, over all time, including paused and archived:
- 1,281 keywords, $2,277.91 spent, 2,821 clicks, zero orders
- 4.6% of everything ever spent on ads
Every one of those dollars was ads not being worth it. The money was small per keyword, about $1.78 each, which is exactly why it was never caught by hand. The account stopped bleeding it when a rule started pausing any keyword with clicks and no orders over 60 days, every day. That is the single most important thing to know about the question in the title: the answer depends on whether something is removing the losers, because Amazon will not.
The ceiling that decides it
Your break-even ACOS is net royalty divided by list price. Two illustrative books, not customers:
| Book | Net royalty | Break-even ACOS | At $4.26 an order |
|---|---|---|---|
| $14.99 paperback, 60%, $4.50 print cost | $4.49 | 30% | Clears by $0.23 |
| $4.99 ebook, 70% | $3.39 | 68% | Loses $0.87 unless read-through or KU covers it |
| $9.99 paperback, 60%, $4.00 print cost | $1.99 | 20% | Loses $2.27 on every ad-attributed sale |
The third book is the trap. Low-priced paperbacks with real print costs have ceilings under 20%, and almost no account holds a 20% ACOS across a catalogue for long (this one averages 25.5%). For that book the honest answer is: not worth it, unless you raise the price or stop advertising it. Run your own books through the break-even ACOS calculator before spending anything.
Two cases where ads look worth it and are not
- Low ACOS on a low-royalty book. A 22% ACOS feels like a win. On the $9.99 paperback above it is a loss on every sale, because the ceiling is 20%. ACOS is only good or bad relative to the royalty, never on its own.
- Attributed sales that belong to other books. Amazon credits any title in your account bought within 14 days of the click to the keyword that got the click. A keyword advertising book one can show a healthy ACOS from sales of book three at a different royalty. If the books have different ceilings, compute the return on the royalties actually earned, not on the ACOS the console shows. The pillar covers this under series books and KENP.
Two cases where ads look bad and are fine
- Launch month ACOS. October 2026 ran at 35.5% ACOS and $6.00 an order, both far worse than the account's averages. It was the right month: new titles with no reviews, bid up on purpose, under the 47% ceiling, $1,566.79 net. Judge a launch at 4 to 6 weeks, not at day 10.
- A KU-heavy book. Attributed sales do not include Kindle Unlimited page reads; those are reported separately as KENP read with an estimated royalty. On a book that earns mostly through KU, the console ACOS overstates the true cost, sometimes badly. Add the KENP royalties to the sales before judging.
- A series starter that loses on its own. A $2.99 first book nets about $2.03 and cannot pay for a $4.26 order. If 40% of its buyers go on to a $4.99 book two, and 70% of those to book three, one sale is worth about $4.34 across the series. Measure read-through; if it is there, the "losing" book is the best ad you have.
How long before you know
Amazon attributes sales for up to 14 days after a click, so nothing is final for two weeks. A keyword needs around 10 clicks before zero orders means anything. A book needs about 100 clicks before its clicks-per-order figure is worth believing. At a dollar a click that is a $100 test per book. If you cannot afford the test, you cannot yet afford the ads, and that is a legitimate answer to the question too.
FAQ
What ACOS means I am losing money?
Any ACOS above your break-even (net royalty / list price). For a $14.99 paperback with a $4.50 print cost that is 30%; for a $4.99 ebook it is about 68%. There is no universal number.
Do Amazon Ads work for a single standalone novel?
They can if the royalty ceiling is above roughly 35% and the page converts, which usually means an ebook at $3.99 or more with reviews. A $2.99 standalone with no read-through rarely clears the cost of an order.
How much did the author behind these numbers waste?
$2,277.91 on 1,281 keywords, found with one filter and now prevented by one rule. That figure is why InteliAds exists; the account above is the founder's, and these pages update as it does.
Sources: InteliAds dashboard on the founder's KDP account; account overview 1 October 2023 to 31 October 2026 read 29 October 2026; targeting view, all marketplaces, all time, all states, filtered to spend with zero orders; monthly views December 2025 and October 2026. One account. Royalty examples are illustrative.